{"id":5016,"date":"2024-06-25T05:48:00","date_gmt":"2024-06-24T20:48:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5016"},"modified":"2024-06-25T05:48:00","modified_gmt":"2024-06-24T20:48:00","slug":"muirbench","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5016","title":{"rendered":"MuirBench"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>MuirBench: A Comprehensive Benchmark for Robust Multi-image Understanding\u00a0<\/strong>[150.3]<br>\u30de\u30eb\u30c1\u30e2\u30fc\u30c0\u30ebLLM\u306e\u5805\u7262\u306a\u30de\u30eb\u30c1\u30a4\u30e1\u30fc\u30b8\u7406\u89e3\u6a5f\u80fd\u306b\u7740\u76ee\u3057\u305f\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u3067\u3042\u308bMuirBench\u3092\u7d39\u4ecb\u3059\u308b\u3002 MuirBench\u306f\u300112\u306e\u591a\u69d8\u306a\u30de\u30eb\u30c1\u30a4\u30e1\u30fc\u30b8\u30bf\u30b9\u30af(\u4f8b\u3048\u3070\u3001\u30b7\u30fc\u30f3\u7406\u89e3\u3001\u9806\u5e8f\u4ed8\u3051)\u3067\u69cb\u6210\u3055\u308c\u3066\u304a\u308a\u300110\u306e\u30ab\u30c6\u30b4\u30ea\u306e\u30de\u30eb\u30c1\u30a4\u30e1\u30fc\u30b8\u95a2\u4fc2\u3092\u542b\u3093\u3067\u3044\u308b\u3002 GPT-4o\u3084Gemini Pro\u306e\u3088\u3046\u306a\u6700\u9ad8\u306e\u30d1\u30d5\u30a9\u30fc\u30de\u30f3\u30b9\u30e2\u30c7\u30eb\u3067\u3055\u3048\u3001\u30e0\u30a2\u30d9\u30f3\u30c1\u3092\u89e3\u304f\u3053\u3068\u306f\u56f0\u96e3\u3067\u3042\u308a\u3001\u7cbe\u5ea6\u306f68.0%\u300149.3%\u3067\u3042\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2406.09411v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2406.09411v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Thu, 13 Jun 2024 17:59:52 GMT)<\/li>\n\n\n\n<li>\u30de\u30eb\u30c1\u30a4\u30e1\u30fc\u30b8\u7406\u89e3\u306e\u305f\u3081\u306e\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3002\u5b9f\u7528\u4e0a\u306f\u5272\u3068\u3088\u304f\u3042\u308a\u305d\u3046\u306a\u72b6\u6cc1\u3060\u304c\u3001\u300cEvaluated upon 20 recent multi-modal LLMs, our results reveal that even the best-performing models like GPT-4o and Gemini Pro find it challenging to solve MUIRBENCH, achieving 68.0% and 49.3% in accuracy.\u300d\u3068\u8aac\u304f\u306e\u304c\u96e3\u3057\u3044\u3088\u3046\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/huggingface.co\/datasets\/MUIRBENCH\/MUIRBENCH\">MUIRBENCH\/MUIRBENCH \u00b7 Datasets at Hugging Face<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[491,524],"class_list":["post-5016","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-491","tag-524"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5016","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=5016"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5016\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5016"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5016"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5016"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}